I definitely have to call left-field on that one. With some dedicated testing and time documenting (with more help than just splittingred and myself), it could be ready for distribution in 2 to 4 weeks. Why/how you ask? I’m already using it on some projects in various forms, I just have not had any time to work on it in the past few weeks due to other commitments; it’s honestly just lacking some usability tweaks, an add-on repository, a few upgrade/migration scripts and some tutorials. I’m not saying every feature on the roadmap is completely fleshed-out, but it’s basically ready, the code is fairly well documented, and, well, it’s working quite well for me.
Thanks for the reply. I had a senior dev evaluate aspects of the code. Conservative estimates are that 0.9.7 is most likely some 12 - 18 months from completion. (Do let me know if you find that completely unreasonable and why. Public or private.) I think therefore our goals will be to work with 0.9.6 and, to the degree possible, aim for upgrade-ability.
I advised SeanPH on the status of 0.9.7, so I should have my behind handed to me. First, however, some clarification:
Quote from: SeanPH at Jan 17, 2008, 04:59 AMWith some dedicated testing and time documenting (with more help than just splittingred and myself), it could be ready for distribution in 2 to 4 weeks. [...] it’s honestly just lacking some usability tweaks, an add-on repository, a few upgrade/migration scripts and some tutorials. [...]basically ready, the code is fairly well documented, and, well, it’s working quite well for me.
[...]Conservative estimates are that 0.9.7 is most likely some 12 - 18 months from completion. (Do let me know if you find that completely unreasonable and why. ) [...]
Perhaps you can enlighten me to the points leading to this conservative estimate?
I agree the public forum discussions in the last 6 to 8 months has stagnated (though we’re very active in team IRC channels), but I’m a little confused on how the repository (I assume you mean the SVN repository?) has stagnated; if anything commits on the 0.9.7 branches accelerated to it’s highest level ever before Christmas. See the changelog at http://svn.modxcms.com/crucible/changelog/MODx/branches/0.9.7
1) Limited activity for the past 6-8 months (slightly picking up closer to the end of the year) - both in the repository and on the discussion board (a dozen threads in the "MODx Next" section). A reasonable chunk of all changes were for 0.9.6. The apparent stagnation can be alarming, especially considering how long the first pre-announcement of 0.9.7 features was made.
I agree there is some cleanup work and bug fixing left to do in many of these areas, but I would hardly call it significant. If you want to discuss specific items there in more details, let’s start some threads on those topics and get some momentum going both in dialog and contributions.
2) Some big-ticket items still require a significant amount of work:
a) user management and security system -
b) packaging/distribution system (at least the manager/interface part)
c) WYSIWYS editor support - I could not get TinyMCE to work, and my understanding is that other editors aren’t supported yet
d) visual design for the management section - things like Ext-JS implementation (which currently uses the humongous ext-all.js), and basic skinning/styling. I believe there recently was a volunteer for look after this.
e) multi-lingual functionality - does not appear to be fully supported yet
f) several internal bugs (e.g. TVs - template variables - get reset if you mistakenly turn a page into a container and then undo the change)
Help me out here with some substance; if this is from your own experience trying to adapt one, where are your posts/questions about the struggle? Seriously, I’m not trying to be smart here, but without the dialog of capable early adopters, progress can be slow and hurdles for other adopters remain too long. Anyway, this is truly something that a quick tutorial on the new xPDO and core MODx API followed by a migration guide with examples of how we converted the major snippets should address quite easily. Where are my tutorial contributors/technical writers?
3) Plugins/modules/extensions/etc will take a while to catch up with the new structure - a few of pro-active projects are being updated already, but this seems to be an exception
100% <expletive removed> agreed here. Where are the other people that want the commodity, really like the potential of the code we’re making available for free, and don’t mind making a bit of an investment in terms of sharing their coding skills to help make it all come together?
4) Two major developers/leads with full time responsibilities elsewhere - brilliant guys, no doubt, but they are still humans who like to sleep and eat occasionally.
I don’t disagree that large migration projects will likely require some significant regression testing, but I’ve been running sites like xpdo.org and other client sites constructed natively on the new core engine for over a year with no significant defects or nasty side-effects. That said, more testing contributions with more feedback in the forums or on the Trac site would be absolutely welcome at this point; every contribution helps.
5) A major overhaul in the underlying structure means some significant testing is needed. Perhaps it’s best to wait for the stable 0.9.7.1 or at least for several weeks post-release?
Look at Ryan’s profile; he just likes to fluff that post count. Feel free to ignore him when he responds to questions about release dates. LOL!
6) The only official word on the release date (up to this post) was "*definitely* before [Oct ’08]"
The new security interfaces need some work and explanation, but the new permission handling that was just introduced in December is quite capable and functional, just quite different. Again, minimal documentation efforts and migration strategies/scripts for the web/manager user convergence are in order, but hardly a show-stopper to product release IMHO.
Only a few of these can be considered a show-stopper (i.e. user/permission handling)
Where are my tutorial contributors/technical writers?Darn it, Jason, I’m working on it! I can’t write about what I don’t understand, and this is taking me way too long to get my head wrapped around.
It is true that the activity picked up in the last couple months of ’07 - my comment tried, unsuccessfully, to point out that there was little activity for the second half of the year as a whole, and most of it occurred close to the end of that ’second half’. Moot point. Glad to hear that IRC is being used for communication - I did not think to check that avenue.
I agree the public forum discussions in the last 6 to 8 months has stagnated (though we’re very active in team IRC channels), but I’m a little confused on how the repository (I assume you mean the SVN repository?) has stagnated; if anything commits on the 0.9.7 branches accelerated to it’s highest level ever before Christmas.
Valid point - specific issues logged within trac will be of far more value that any generic summaries I could post here. To say nothing of also logging patches for these issues.
I agree there is some cleanup work and bug fixing left to do in many of these areas, but I would hardly call it significant. If you want to discuss specific items there in more details, let’s start some threads on those topics and get some momentum going both in dialog and contributions.
Certainly. One example is Ditto - it seems to be one of the most popular snippets, and yet it does not work out of the box with 097 (if _I_ messed something up, then I’ll be quick to retract this statement). "does not work", in this context, is useless, but honestly I was not in the error-solving mode when I tried to use it; it was more a quick mission to find out what I can get working as quickly as possible and Ditto did not make the cut.
Help me out here with some substance; if this is from your own experience trying to adapt one, where are your posts/questions about the struggle? Seriously, I’m not trying to be smart here, but without the dialog of capable early adopters, progress can be slow and hurdles for other adopters remain too long. Anyway, this is truly something that a quick tutorial on the new xPDO and core MODx API followed by a migration guide with examples of how we converted the major snippets should address quite easily. Where are my tutorial contributors/technical writers?
Quote from: e6 at Jan 18, 2008, 02:06 AM100% <expletive removed> agreed here. Where are the other people that want the commodity, really like the potential of the code we’re making available for free, and don’t mind making a bit of an investment in terms of sharing their coding skills to help make it all come together?
4) Two major developers/leads with full time responsibilities elsewhere - brilliant guys, no doubt, but they are still humans who like to sleep and eat occasionally.
???
This is actually much further along than the ticket suggests. It really just needs some UI enhancements, maybe a bit of inline documentation, and some testing.
2) Some big-ticket items still require a significant amount of work:
a) user management and security system -
I would agree this needs some work. We’ve come a long way on it - the core code is pretty much done (minus the provisioners part). I would definitely agree that the UI needs work.
b) packaging/distribution system (at least the manager/interface part)
TinyMCE works a bit different in 097 - you have to install it as a package. I’ve gotten it to work on mine - and in fact, once you get it to work, you’ll be able to see the sweet new MODx.Browser that replaces the mcpuk file browser.
c) WYSIWYS editor support - I could not get TinyMCE to work, and my understanding is that other editors aren’t supported yet
Yes, such is the debate. I’ve made significant progress in porting some ext1.1.1 code to ext2, so there’s a chance (read: small chance) we might go to ext2 for 097. We’ll see.
d) visual design for the management section - things like Ext-JS implementation (which currently uses the humongous ext-all.js), and basic skinning/styling. I believe there recently was a volunteer for look after this.
It’s very near done - in fact, it’s just a matter of including some preference options and a hook here and there. Jason can explain more.
e) multi-lingual functionality - does not appear to be fully supported yet
Logging these bugs to Trac would be infinitely helpful.
f) several internal bugs (e.g. TVs - template variables - get reset if you mistakenly turn a page into a container and then undo the change)
I agree - we’ve already done a lot of work on some plugins, in our modx-components repository. I’ve gotten significant work done on AjaxSearch and eForm, converting them into OOP and using the xPDO core. They still need work, but they’re there.
3) Plugins/modules/extensions/etc will take a while to catch up with the new structure - a few of pro-active projects are being updated already, but this seems to be an exception
Oh, I like to sleep and eat FAR more than occasionally.
4) Two major developers/leads with full time responsibilities elsewhere - brilliant guys, no doubt, but they are still humans who like to sleep and eat occasionally.
Actually, surprisingly, we’re doing pretty good testing-wise. We’ve got quite a few sites running on 0.9.7, and running well. Sure, there are definitely going to be testing needs, but a big reason 097 has taken so long is that we’re doing a ton of testing as we code, making sure that the code is 100% compliant (XHTML and JS standards) and written in proper OOP style.
5) A major overhaul in the underlying structure means some significant testing is needed. Perhaps it’s best to wait for the stable 0.9.7.1 or at least for several weeks post-release?
Only a few of these can be considered a show-stopper (i.e. user/permission handling), but aren’t these issues enough to estimate 0.9.7 as being considerably more than 4 weeks away, especially as a solid, tested version?That’s why it’d be an alpha release.
No - please do. These posts are wonderful, and vastly helpful for the project. I’ve really appreciated the feedback.
Whatever the answer is, you guys are doing a truly fantastic job - don’t let us nitpicking bystanders interrupt you too much